通过输入澄清集成分解大语言模型的不确定性
Decomposing Uncertainty for Large Language Models through Input Clarification Ensembling
AI总结:
本文提出输入澄清集成框架,通过生成输入澄清并集成预测,将LLM总不确定性分解为偶然和认知两部分,在多个语言任务上实现可靠的不确定性量化。
摘要:
不确定性分解是指将预测模型的总不确定性分解为偶然(数据)不确定性和认知(模型)不确定性的任务,前者源于数据生成过程中的固有随机性,后者源于模型训练数据中缺失的信息。特别是在大语言模型(LLMs)中,识别不确定性来源是提高可靠性、可信度和可解释性的重要一步,但仍然是一个重要的开放研究问题。在本文中,我们为LLMs引入了一个不确定性分解框架,称为输入澄清集成(input clarification ensembling),该框架可应用于任何预训练的LLM。我们的方法为输入生成一组澄清,将其输入LLM,并对相应的预测进行集成。我们表明,当偶然不确定性源于LLM输入中的歧义或欠指定时,该方法使得能够使用类似于贝叶斯神经网络所采用的分解方式,将(未澄清的)LLM的预测分解为独立的偶然项和认知项。实证评估表明,输入澄清集成在多个语言处理任务上提供了准确且可靠的不确定性量化。代码和数据可在 https://github.com/UCSB-NLP-Chang/llm_uncertainty 获取。
英文摘要:
Uncertainty decomposition refers to the task of decomposing the total uncertainty of a predictive model into aleatoric (data) uncertainty, resulting from inherent randomness in the data-generating process, and epistemic (model) uncertainty, resulting from missing information in the model's training data. In large language models (LLMs) specifically, identifying sources of uncertainty is an important step toward improving reliability, trustworthiness, and interpretability, but remains an important open research question. In this paper, we introduce an uncertainty decomposition framework for LLMs, called input clarification ensembling, which can be applied to any pre-trained LLM. Our approach generates a set of clarifications for the input, feeds them into an LLM, and ensembles the corresponding predictions. We show that, when aleatoric uncertainty arises from ambiguity or under-specification in LLM inputs, this approach makes it possible to factor an (unclarified) LLM's predictions into separate aleatoric and epistemic terms, using a decomposition similar to the one employed by Bayesian neural networks. Empirical evaluations demonstrate that input clarification ensembling provides accurate and reliable uncertainty quantification on several language processing tasks. Code and data are available at https://github.com/UCSB-NLP-Chang/llm_uncertainty.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。